[5.0.x] Fixed CVE-2024-41991 -- Prevented potential ReDoS in django.utils.html.urlize() and AdminURLFieldWidget.

Thanks Seokchan Yoon for the report.

Co-authored-by: Sarah Boyce <42296566+sarahboyce@users.noreply.github.com>

django/contrib/admin/widgets.py

@@ -390,7 +390,7 @@ class AdminURLFieldWidget(forms.URLInput):
         context["current_label"] = _("Currently:")
         context["change_label"] = _("Change:")
         context["widget"]["href"] = (
-            smart_urlquote(context["widget"]["value"]) if value else ""
+            smart_urlquote(context["widget"]["value"]) if url_valid else ""
         )
         context["url_valid"] = url_valid
         return context

django/utils/html.py

@@ -36,6 +36,8 @@ VOID_ELEMENTS = {
     "spacer",
 }
 
+MAX_URL_LENGTH = 2048
+
 
 @keep_lazy(SafeString)
 def escape(text):
@@ -330,9 +332,9 @@ class Urlizer:
             # Make URL we want to point to.
             url = None
             nofollow_attr = ' rel="nofollow"' if nofollow else ""
-            if self.simple_url_re.match(middle):
+            if len(middle) <= MAX_URL_LENGTH and self.simple_url_re.match(middle):
                 url = smart_urlquote(html.unescape(middle))
-            elif self.simple_url_2_re.match(middle):
+            elif len(middle) <= MAX_URL_LENGTH and self.simple_url_2_re.match(middle):
                 url = smart_urlquote("http://%s" % html.unescape(middle))
             elif ":" not in middle and self.is_email_simple(middle):
                 local, domain = middle.rsplit("@", 1)
@@ -447,6 +449,10 @@ class Urlizer:
         except ValueError:
             # value contains more than one @.
             return False
+        # Max length for domain name labels is 63 characters per RFC 1034.
+        # Helps to avoid ReDoS vectors in the domain part.
+        if len(p2) > 63:
+            return False
         # Dot must be in p2 (e.g. example.com)
         if "." not in p2 or p2.startswith("."):
             return False

docs/releases/4.2.15.txt

@@ -23,6 +23,13 @@ CVE-2024-41990: Potential denial-of-service vulnerability in ``django.utils.html
 denial-of-service attack via very large inputs with a specific sequence of
 characters.
 
+CVE-2024-41991: Potential denial-of-service vulnerability in ``django.utils.html.urlize()`` and ``AdminURLFieldWidget``
+=======================================================================================================================
+
+:tfilter:`urlize`, :tfilter:`urlizetrunc`, and ``AdminURLFieldWidget`` were
+subject to a potential denial-of-service attack via certain inputs with a very
+large number of Unicode characters.
+
 Bugfixes
 ========
 

docs/releases/5.0.8.txt

@@ -23,6 +23,13 @@ CVE-2024-41990: Potential denial-of-service vulnerability in ``django.utils.html
 denial-of-service attack via very large inputs with a specific sequence of
 characters.
 
+CVE-2024-41991: Potential denial-of-service vulnerability in ``django.utils.html.urlize()`` and ``AdminURLFieldWidget``
+=======================================================================================================================
+
+:tfilter:`urlize`, :tfilter:`urlizetrunc`, and ``AdminURLFieldWidget`` were
+subject to a potential denial-of-service attack via certain inputs with a very
+large number of Unicode characters.
+
 Bugfixes
 ========
 

tests/admin_widgets/tests.py

@@ -462,7 +462,12 @@ class AdminSplitDateTimeWidgetTest(SimpleTestCase):
 class AdminURLWidgetTest(SimpleTestCase):
     def test_get_context_validates_url(self):
         w = widgets.AdminURLFieldWidget()
-        for invalid in ["", "/not/a/full/url/", 'javascript:alert("Danger XSS!")']:
+        for invalid in [
+            "",
+            "/not/a/full/url/",
+            'javascript:alert("Danger XSS!")',
+            "http://" + "한.글." * 1_000_000 + "com",
+        ]:
             with self.subTest(url=invalid):
                 self.assertFalse(w.get_context("name", invalid, {})["url_valid"])
         self.assertTrue(w.get_context("name", "http://example.com", {})["url_valid"])

tests/utils_tests/test_html.py

@@ -338,6 +338,15 @@ class TestUtilsHtml(SimpleTestCase):
                 'Search for <a href="http://google.com/?q=">google.com/?q=</a>!',
             ),
             ("foo@example.com", '<a href="mailto:foo@example.com">foo@example.com</a>'),
+            (
+                "test@" + "한.글." * 15 + "aaa",
+                '<a href="mailto:test@'
+                + "xn--6q8b.xn--bj0b." * 15
+                + 'aaa">'
+                + "test@"
+                + "한.글." * 15
+                + "aaa</a>",
+            ),
         )
         for value, output in tests:
             with self.subTest(value=value):
@@ -346,6 +355,10 @@ class TestUtilsHtml(SimpleTestCase):
     def test_urlize_unchanged_inputs(self):
         tests = (
             ("a" + "@a" * 50000) + "a",  # simple_email_re catastrophic test
+            # Unicode domain catastrophic tests.
+            "a@" + "한.글." * 1_000_000 + "a",
+            "http://" + "한.글." * 1_000_000 + "com",
+            "www." + "한.글." * 1_000_000 + "com",
             ("a" + "." * 1000000) + "a",  # trailing_punctuation catastrophic test
             "foo@",
             "@foo.com",